注意力卷积GRU自编码器及其在工业过程监控的应用
|
刘兴,余建波
|
Attention convolutional GRU-based autoencoder and its application in industrial process monitoring
|
Xing LIU,Jian-bo YU
|
|
表 5 CGRUA-AE与深度学习检测方法的TEP故障FDR/DR |
Tab.5 FDR/DR of CGURA-AE and deep learning detection methods on TEP |
|
方法 | 故障类型 | 平均值 | 阶跃 | 随机变量 | 未知 | 其他 | CNN | 0.02/0.86 | 0.03/0.79 | 0.02/0.68 | 0.03/0.80 | 0.025/0.785 | LSTM | 0.02/0.94 | 0.03/0.79 | 0.02/0.67 | 0.02/0.80 | 0.019/0.809 | SDAE (T2) | 0.02/0.66 | 0.04/0.60 | 0.03/0.45 | 0.04/0.74 | 0.028/0.602 | SDAE(SPE) | 0.03/0.88 | 0.05/0.87 | 0.09/0.73 | 0.07/0.85 | 0.061/0.832 | DBN (T2) | 0.01/0.87 | 0.01/0.79 | 0.02/0.55 | 0.02/0.81 | 0.014/0.75 | DBN(SPE) | 0.01/0.98 | 0.01/0.82 | 0.01/0.78 | 0.01/0.79 | 0.011/0.856 | CGRUA-AE (T2) | 0.02/1 | 0.03/0.89 | 0.02/0.77 | 0.02/0.82 | 0.023/0.879 | CGRUA-AE(SPE) | 0.02/0.99 | 0.03/0.88 | 0.02/0.86 | 0.01/0.81 | 0.02/0.903 |
|
|
|